[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20081217113804F.fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 17 Dec 2008 11:48:41 +0900
From: FUJITA Tomonori <fujita.tomonori@....ntt.co.jp>
To: jeremy@...p.org
Cc: mingo@...e.hu, linux-kernel@...r.kernel.org,
xen-devel@...ts.xensource.com, x86@...nel.org,
ian.campbell@...rix.com, jbeulich@...ell.com,
fujita.tomonori@....ntt.co.jp
Subject: Re: [PATCH 04 of 14] swiotlb: consistently use
address_needs_mapping everywhere
On Tue, 16 Dec 2008 12:17:28 -0800
Jeremy Fitzhardinge <jeremy@...p.org> wrote:
> Signed-off-by: Ian Campbell <ian.campbell@...rix.com>
> ---
> lib/swiotlb.c | 10 +++-------
> 1 file changed, 3 insertions(+), 7 deletions(-)
This patch is wrong.
dma_mask and coherent_dma_mask represent different restrictions.
You can't use coherent_dma_mask for alloc_coherent.
> diff --git a/lib/swiotlb.c b/lib/swiotlb.c
> --- a/lib/swiotlb.c
> +++ b/lib/swiotlb.c
> @@ -465,13 +465,9 @@
> dma_addr_t dev_addr;
> void *ret;
> int order = get_order(size);
> - u64 dma_mask = DMA_32BIT_MASK;
> -
> - if (hwdev && hwdev->coherent_dma_mask)
> - dma_mask = hwdev->coherent_dma_mask;
>
> ret = (void *)__get_free_pages(flags, order);
> - if (ret && !is_buffer_dma_capable(dma_mask, virt_to_bus(ret), size)) {
> + if (ret && !address_needs_mapping(hwdev, virt_to_bus(ret), size)) {
> /*
> * The allocated memory isn't reachable by the device.
> * Fall back on swiotlb_map_single().
> @@ -495,9 +491,9 @@
> dev_addr = virt_to_bus(ret);
>
> /* Confirm address can be DMA'd by device */
> - if (!is_buffer_dma_capable(dma_mask, dev_addr, size)) {
> + if (!address_needs_mapping(hwdev, dev_addr, size)) {
> printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n",
> - (unsigned long long)dma_mask,
> + (unsigned long long)dma_get_mask(hwdev),
> (unsigned long long)dev_addr);
>
> /* DMA_TO_DEVICE to avoid memcpy in unmap_single */
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists